{"name":"sam3-segmentation","repo":"lillyguisnet/sam3-segmentation","url":"https://midorreal.com/project/lillyguisnet-sam3-segmentation","repository":"https://github.com/lillyguisnet/sam3-segmentation","description":"Agent skill: outline, mask and count objects in photos with Meta's SAM 3 on your own NVIDIA GPU, with review, exports and a viewer. Works with Claude Code, Codex, Cursor, Pi and other Agent Skills clients.","facts":{"license":{"value":"MIT","source":"https://github.com/lillyguisnet/sam3-segmentation","checked":"22 Nov 2026"},"language":{"value":"Python","source":"https://github.com/lillyguisnet/sam3-segmentation","checked":"22 Nov 2026"},"last_release":{"value":"v0.1.0, 24 Oct 2026","source":"https://github.com/lillyguisnet/sam3-segmentation/releases","checked":"22 Nov 2026"},"releases":{"value":"1","source":"https://github.com/lillyguisnet/sam3-segmentation/releases","checked":"20 Nov 2026"},"contributors":{"value":"1","source":"https://github.com/lillyguisnet/sam3-segmentation/graphs/contributors","checked":"20 Nov 2026"},"archived":{"value":"no","source":"https://github.com/lillyguisnet/sam3-segmentation","checked":"22 Nov 2026"}},"stars":149.0,"writeup":{"what_it_is":"An agent skill that uses Meta's SAM 3 model, run on the user's own NVIDIA GPU, to outline, mask and count objects in photos. It includes review steps, exports and a viewer, and works with several coding agents.","audience":"Developers and agent users who need to segment or count objects in photos on an NVIDIA GPU","claims":[{"kind":"specific","claim":"Requires an NVIDIA GPU with 12-24 GB memory suggested","excerpt":"an **NVIDIA GPU** (12-24 GB memory suggested)","status":"not_checked"},{"kind":"specific","claim":"Requires about 11 GB of disk space","excerpt":"about **11 GB** of disk","status":"not_checked"},{"kind":"specific","claim":"Includes a demo that runs without a GPU, using generated shapes","excerpt":"Any machine with [`uv`](https://docs.astral.sh/uv/getting-started/installation/); no GPU, model or account","status":"not_checked"},{"kind":"broad","claim":"Described as easy to use with coding agents","excerpt":"this skill gives it the model, the commands and the rules to do it properly","status":"too_broad"}],"alternatives":[],"written_by":"AI, from the project's README","date":"2026-10-28"},"why_now":null,"owner_supplied":null,"score":{"verdict":"too_early","hype":54,"reality":14,"gap":39,"momentum":null,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T08:00:00+00:00","verdict_source":"rule"},"early_read":{"estimate":true,"lean":"too_close","gap_low":0,"gap_high":40,"rule":"early-2.0"},"changes":[{"date":"2026-10-24T01:52:52+00:00","text":"Latest release v0.1.0","source":"https://github.com/lillyguisnet/sam3-segmentation/releases"},{"date":"2026-10-24T01:49:15+00:00","text":"Repository created","source":"https://github.com/lillyguisnet/sam3-segmentation"}],"cite":"sam3-segmentation is a GitHub project at https://github.com/lillyguisnet/sam3-segmentation written mainly in Python, described there as \"Agent skill: outline, mask and count objects in photos with Meta's SAM 3 on your own NVIDIA GPU, with review, exports and a viewer. Works with Claude Code, Codex, Cursor, Pi and other Agent Skills clients\".","attribution":"Data from Mid or Real, https://midorreal.com/project/lillyguisnet-sam3-segmentation, checked 22 November 2026. Attribution with a link is required.","still_real":null}